8-TQEN (N,N,N′,N′-tetrakis(8-quinolylmethyl)ethylenediamine) analogs as fluorescent cadmium sensors: strategies to enhance Cd²⁺-induced fluorescence and Cd²⁺/Zn²⁺ selectivity

Abstract

In order to exploit the untapped sensing potential of the TQEN (N,N,N′,N′-tetrakis(2-quinolylmethyl)ethylenediamine) family of fluorescent probes, 8-TQEN (N,N,N′,N′-tetrakis(8-quinolylmethyl)ethylenediamine) analogs were designed and characterized. Although 8-TQEN lacks practicality owing to poor solubility in both aqueous media and organic solvents, 6-MeO-8-TQEN (N,N,N′,N′-tetrakis(6-methoxy-8-quinolylmethyl)ethylenediamine, 1b) exhibits Cd²⁺-selective fluorescence enhancement at 395 nm in DMF–HEPES buffer (1 : 1), (I_Cd/I₀ = 25, λ_ex = 332 nm, ϕ_Cd = 0.025). Zn²⁺ induces weaker fluorescence (I_Zn/I₀ = 9, I_Zn/I_Cd = 0.35). In contrast, both the parent probes TQEN (I_Zn/I₀ = 23, I_Cd/I_Zn = 0.64) and 6-MeOTQEN (I_Zn/I₀ = 11, I_Cd/I_Zn = 1.29 exhibit higher sensitivity toward Zn²⁺. When the quinoline groups were replaced with 8-hydroxyquinoline different responses were observed. The propanediamine derivative, 8-TQOEPN (N,N,N′,N′-tetrakis(8-quinolyloxyethylene)propanediamine, 2c) exhibits significant fluorescence enhancement at 423 nm upon Cd²⁺ binding (λ_ex = 325 nm, I_Cd/I₀ = 19, ϕ_Cd = 0.31). Fluorescence enhancement is Cd²⁺-specific as Zn²⁺ induces only more modest emission increases (I_Zn/I₀ = 8.8, I_Zn/I_Cd = 0.45).

---

Introduction

Techniques to visualize specific metal ions are necessary to investigate inorganic biochemistry and environmental pollutants.^1–6 Since metal ions perform indispensable tasks in biological systems, metal ion concentration in organs and/or whole cells is tightly regulated. To maintain optimal function, cells must acquire and excrete metal ions efficiently to prevent both deficiencies and excesses in concentrations. Despite the essentiality of regulation, the exact metal ion homeostasis mechanisms are still being elucidated. Fluorescent probes are an attractive approach to visualize metal ions for numerous applications.^2,7 Fluorescent Zn²⁺ sensors have been investigated extensively to interrogate the biological roles of zinc.^8–11 The heavy elements from group 12, Hg²⁺ and Cd²⁺, are also potential targets for fluorescent sensor development. New Hg²⁺ and Cd²⁺ probes could facilitate investigating the mechanisms of toxicity as well as therapeutic strategies.^12–17 Since Zn²⁺ and Cd²⁺ possess similar coordination chemistry, discriminating between the two metal ions is a significant and important challenge in sensor development.^6,18–22

One of our long-standing goals is to develop structurally simple, Cd²⁺-specific fluorescent sensors that can be made readily available to the scientific community. Previously, we reported that TQEN (N,N,N′,N′-tetrakis(2-quinolylmethyl)ethylenediamine), a simple analog of TPEN (N,N,N′,N′-tetrakis(2-pyridylmethyl)ethylenediamine), exhibits emission enhancement upon Zn²⁺ complexation (Chart 1).²³ The emission enhancement observed for TQEN results from a combination of the PET (photo-induced electron transfer) and the CHEF (chelation enhanced fluorescence) mechanisms of fluorescence switching.^1,6 In addition to Zn-induced emission enhancement, TQEN also responds to Cd²⁺ (I_Cd/I_Zn = 0.64). The introduction of 6-methoxy group onto quinoline ring of TQEN increases the Cd²⁺ sensitivity in 6-MeOTQEN (I_Cd/I_Zn = 1.29).²⁴ Since the substitution of quinoline for the isoquinoline chromophore^25,26 and the modification of molecular scaffold^21,26 reduced the Cd²⁺ response significantly, we sought an alternative strategy to increase Cd²⁺-selectivity.

Chart 1

In this paper, we describe the Cd²⁺-induced fluorescence enhancement of 6-MeO-8-TQEN (N,N,N′,N′-tetrakis(6-methoxy-8-quinolylmethyl)ethylenediamine, 1b, Chart 2) and 8-TQOEPN (N,N,N′,N′-tetrakis(2-(8-quinolyloxy)ethyl)propanediamine), 2c, Chart 2). Isomers with different chain lengths also were investigated to further examine the structural requirements need to obtain an efficient fluorescence response.

Chart 2

Experimental

General

All reagents and solvents used for synthesis were from commercial sources and used as received. Compound 2a is a known compound²⁷ and was synthesized from a chloride precursor.^28,29 The other compounds were prepared as outlined in Scheme 1 from reported compounds.^29,30 N,N-Dimethylformamide (DMF, Dojin) was spectral grade (Spectrosol). All aqueous solution was prepared using Milli-Q water (Millipore). ¹H NMR (300 Hz) and ¹³C NMR (75.5 Hz) spectra were recorded on a Varian GEMINI 2000 spectrometer and referenced to internal Si(CH₃)₄ or solvent signals. UV-vis and fluorescence spectra were measured on a Jasco V-660 spectrophotometer and Jasco FP-6300 spectrofluorometer, respectively. Fluorescence quantum yields were measured on a HAMAMATSU photonics C9920-02 absolute PL quantum yield measurement system.

Scheme 1

N,N,N′,N′-Tetrakis(8-quinolylmethyl)ethylenediamine (8-TQEN, 1a)

To a dry CH₃CN solution (30 mL) of 8-bromomethylquinoline (1.72 g, 7.75 mmol) and ethylenediamine (0.13 mL, 1.94 mmol) was added potassium carbonate (1.74 g, 12.6 mmol) and potassium iodide (573 mg, 3.45 mmol). The resulting reaction mixture was refluxed for 2 days under N₂. The resultant solution was cooled to room temperature and the solvent was evaporated. The residue was washed with CHCl₃ and water to give compound 1a as a white solid (450 mg, 0.72 mmol, 37%). This material is insoluble to water, ethanol, chloroform, DMF, DMSO and other common organic solvents. Anal. calcd for C₄₂H₃₆N₆ (1a·0.2H₂O): C, 80.29; H, 5.84; N, 13.37. Found: C, 80.07; H, 5.78; N, 13.28.

N,N,N′,N′-Tetrakis(6-methoxy-8-quinolylmethyl)ethylenediamine (6-MeO-8-TQEN, 1b)

To a dry CH₃CN solution (30 mL) of 6-methoxy-8-bromomethylquinoline (118 mg, 0.56 mmol) and ethylenediamine (9.5 μL, 0.14 mmol) was added potassium carbonate (196 mg, 1.40 mmol) and potassium iodide (236 mg, 1.40 mmol). The resulting reaction mixture was refluxed for 2 days under N₂, cooled to room temperature and the solvent was evaporated. The residue was extracted with CHCl₃–water and the organic layer was dried and evaporated to give compound 1b as a white solid (109 mg, 0.14 mmol, quant.). ¹H NMR (DMSO-d₆): δ (ppm) 8.60 (dd, 4H, J = 1.7, 4.1 Hz), 8.17 (dd, 4H, J = 1.7, 8.1 Hz), 7.61 (d, 4H, J = 2.4 Hz), 7.40 (dd, 4H, J = 4.1, 8.1 Hz), 7.12 (d, 4H, J = 2.4 Hz), 4.32 (s, 8H), 3.74 (s, 12H), 2.93 (s, 4H). ¹³C NMR (DMSO-d₆): δ (ppm) 156.7, 146.4, 142.0, 139.1, 134.8, 129.0, 121.3, 119.9, 103.8, 87.7, 55.2, 54.1. HRMS (ESI-MS) m/z: [M + Na]⁺ calcd for C₄₆H₄₄N₆O₄Na 745.3502; found 745.3567. Anal. calcd for C₄₆H₄₅N₆O_4.5 (1b·0.5H₂O): C, 73.28; H, 6.01; N, 11.14. Found: C, 73.23; H, 6.04; N, 11.17.

N,N,N′,N′-Tetrakis(2-(8-quinolyloxy)ethyl)ethylenediamine (8-TQOEEN, 2a)

To an agitated mixture of 8-quinolinol (2.18 g, 14.8 mmol) and powdered potassium hydroxide (830 mg, 14.8 mmol) in dry ethanol (30 mL) was added N,N,N′,N′-tetrakis(2-chloroethyl)ethylenediamine (4a) (460 mg, 1.48 mmol) in 3 mL of ethanol. The resulting reaction mixture was refluxed overnight under N₂, cooled to room temperature and the solvent was evaporated. The residue was extracted with CHCl₃–water and the organic layer was washed with 3 N NaOH, dried and evaporated. The residue was purified by column chromatography (alumina, CHCl₃/CH₃OH 99.5/0.5, R_f = 0.35) to give compound 2a as a yellow oil (226 mg, 0.30 mmol, 20%). ¹H NMR (CDCl₃): δ (ppm) 8.91 (dd, J = 4.3, 1.8 Hz, 4H), 8.14 (dd, J = 8.2, 1.5 Hz, 4H), 7.40–7.45 (m, 12H), 7.06 (dd, J = 7.3, 1.5 Hz, 4H), 4.34 (t, J = 5.8 Hz, 8H), 3.05 (t, J = 6.0 Hz, 8H), 2.76 (s, 4H). ¹³C NMR (CDCl₃): δ (ppm) 154.0, 148.6, 139.5, 136.0, 129.3, 126.5, 121.6, 119.5, 64.2, 56.8, 52.5. HRMS (ESI-MS) m/z: [M + Na]⁺ calcd for C₄₆H₄₄N₆O₄Na 767.3322; found 767.3434.

N,N,N′,N′-Tetrakis(3-(8-quinolyloxy)propyl)ethylenediamine (8-TQOPEN, 2b)

To an agitated mixture of 8-quinolinol (1.88 g, 13.0 mmol) and powdered potassium hydroxide (788 mg, 14.0 mmol) in dry ethanol (30 mL) was added N,N,N′,N′-tetrakis(3-chloropropyl)ethylenediamine (4b) (397 mg, 1.08 mmol) in 3 mL of ethanol. The resulting reaction mixture was refluxed overnight under N₂, cooled to room temperature and the solvent was evaporated. The residue was extracted with CHCl₃–water and the organic layer was washed with 3 N NaOH, dried and evaporated. The residue was purified by column chromatography (alumina, CHCl₃/CH₃OH 99.5/0.5, R_f = 0.28) to give a yellow oil. This material was dissolved in small portions of acetonitrile and conc. hydrochloric acid was added to induce precipitation. After filtration, the obtained hydrochloride salt was dissolved in water and 6 N NaOH was added to adjust the pH > 12. The product was extracted into chloroform, dried and evaporated to give compound 2b as a yellow oil (130 mg, 0.162 mmol, 15%). ¹H NMR (CDCl₃): δ (ppm) 8.89 (dd, J = 4.3, 1.5 Hz, 4H), 8.05 (dd, J = 8.5, 1.5 Hz, 4H), 7.27–7.38 (m, 12H), 6.87 (dd, J = 7.3, 1.5 Hz, 4H), 4.06 (t, J = 6.6 Hz, 8H), 2.55 (t, J = 6.6 Hz, 8H), 2.43 (s, 4H), 1.88 (quint, J = 6.5 Hz, 8H). ¹³C NMR (CDCl₃): δ (ppm) 154.5, 148.9, 140.1, 135.5, 129.2, 126.5, 121.3, 119.1, 108.5, 66.7, 52.6, 50.7, 26.9. HRMS (ESI-MS) m/z: [M + Na]⁺ calcd for C₅₀H₅₂N₆O₄Na 823.3948; found 823.3985. Anal. calcd for C₅₀H₅₇Cl₂N₆O_5.5 (2b·2HCl·1.5H₂O): C, 66.66; H, 6.38; N, 9.33. Found: C, 66.77; H, 5.99; N, 9.23.

N,N,N′,N′-Tetrakis(2-(8-quinolyloxy)ethyl)propanediamine (8-TQOEPN, 2c)

To an agitated mixture of 8-quinolinol (1.53 g, 10.6 mmol) and powdered potassium hydroxide (592 mg, 10.6 mmol) in dry ethanol (30 mL) was added N,N,N′,N′-tetrakis(2-chloroethyl)propanediamine (4c) (285 mg, 1.06 mmol) in 3 mL of ethanol. The resulting reaction mixture was refluxed overnight under N₂, cooled to room temperature and the solvent was evaporated. The residue was extracted with CHCl₃–water and the organic layer was washed with 3 N NaOH, dried and evaporated. The residue was purified by column chromatography (alumina, CHCl₃/CH₃OH 99.5/0.5, R_f = 0.3) to give compound 2c as a yellow oil (88 mg, 0.12 mmol, 9%). ¹H NMR (CDCl₃): δ (ppm) 8.88 (dd, J = 4.3, 1.8 Hz, 4H), 8.06 (dd, J = 8.2, 1.5 Hz, 4H), 7.29–7.40 (m, 12H), 7.02 (dd, J = 6.4, 2.7 Hz, 4H), 4.34 (t, J = 6.6 Hz, 8H), 4.34 (t, J = 6.7 Hz, 8H), 3.24 (t, J = 6.6 Hz, 8H), 2.82 (t, J = 7.2 Hz, 8H), 2.61 (s, 4H), 1.85 (br, 2H). ¹³C NMR (CDCl₃): δ (ppm) 154.2, 148.9, 140.0, 135.5, 129.2, 126.5, 121.3, 119.3, 108.7, 77.2, 67.2, 53.6, 25.4. HRMS (ESI-MS) m/z: [M + Na]⁺ calcd for C₄₇H₄₆N₆O₄Na 781.3398; found 781.3478. Anal. calcd for C₄₇H₅₀N₆O₆ (2c·2H₂O): C, 71.01; H, 6.34; N, 10.57. Found: C, 70.77; H, 6.35; N, 10.88.

N,N,N′,N′-Tetrakis(3-(8-quinolyloxy)propyl)propanediamine (8-TQOPPN, 2d)

To an agitated mixture of 8-quinolinol (2.67 g, 18.4 mmol) and powdered potassium hydroxide (1.03 g, 18.4 mmol) in dry ethanol (30 mL) was added N,N,N′,N′-tetrakis(3-chloropropyl)propanediamine (4d) (700 mg, 1.84 mmol) in 3 mL of ethanol. The resulting reaction mixture was refluxed overnight under N₂. The resultant solution was cooled to room temperature and the solvent was evaporated. The residue was extracted with CHCl₃–water and the organic layer was washed with 3 N NaOH, dried and evaporated. The residue was purified by column chromatography (alumina, AcOEt/MeOH 98/2, R_f = 0.6) to give yellow oil. This material was dissolved in a small portion of ethanol and conc. hydrochloric acid was added to induce precipitation. After filtration, the hydrochloride salt was dissolved in water and 6 N NaOH was added to adjust the pH > 12. The product was extracted into chloroform, dried and evaporated to give compound 2d as a yellow oil (195 mg, 0.239 mmol, 13%). ¹H NMR (CDCl₃): δ (ppm) 8.90 (dd, J = 4.3, 1.8 Hz, 4H), 8.04 (dd, J = 8.2, 1.8 Hz, 4H), 7.26–7.38 (m, 12H), 6.89 (dd, J = 7.3, 1.5 Hz, 4H), 4.10 (t, J = 6.7 Hz, 8H), 2.57 (br, 8H), 2.40 (br, 4H), 2.06 (quint, J = 6.6 Hz, 8H), 1.54 (br, 2H). ¹³C NMR (CDCl₃): δ (ppm) 154.5, 148.8, 140.0, 135.5, 129.1, 126.4, 121.2, 119.0, 108.4, 66.9, 52.3, 50.1, 26.8, 24.5. HRMS (ESI-MS) m/z: [M + H]⁺ calcd for C₅₁H₅₅N₆O₄ 815.4286; found 815.4285. Anal. calcd for C₅₁H₆₀ClN₂O_6.5 (2d·HCl·2.5H₂O): C, 68.32; H, 6.75; N, 9.37. Found: C, 68.51; H, 6.35; N, 9.26.

Results and discussion

Ligand synthesis

Compounds 1a and 1b were synthesized from corresponding 8-bromomethylquinolines and ethylenediamine. Ligands 2b–d were synthesized using the procedures outlined in Scheme 1. The final product yields are relatively low due to difficult purifications. No obvious correlations between the chain lengths and yields were observed. All new compounds were characterized by ¹H and ¹³C NMR, and the purity was verified by elemental analysis.

UV-vis and fluorescence spectral changes of 6-MeO-8-TQEN (1b) induced by Cd²⁺

Since compound 8-TQEN (1a) exhibited poor solubility in common organic solvents, spectroscopic investigations could not be executed. Spectral measurements of 6-MeO-8-TQEN (1b) were performed in DMF–HEPES buffer (1 : 1 (v/v), HEPES buffer contains 50 mM HEPES, 100 mM KCl, pH = 8.0) at 25 °C (Fig. 1). Upon addition of Cd²⁺, the UV-vis absorption of 1b changed slightly, however, no distinct isosbestic points were detected. Fig. 1b shows the emission changes in 1b after the addition of Cd²⁺. Upon excitation at 332 nm, apo-1b exhibits negligible fluorescence. After the addition of Cd²⁺, the fluorescence increases at 395 nm. The fluorescence quantum yield of [Cd(1b)]²⁺ (ϕ = 0.025) is similar to [Zn(6-MeOTQEN)]²⁺ (ϕ = 0.032) and larger than [Zn(TQEN)]²⁺ (ϕ = 0.007). From the fluorescence titration curve shown in Fig. 1c, the dissociation constant of [Cd(1b)]²⁺ complex was calculated to be K_d = (3.4 ± 0.1) × 10⁻⁵ M using nonlinear curve fitting (Fig. S1†).²⁴ The binding affinity is significantly weaker than that of the [Zn(TQEN)]²⁺ complex (K_d = ∼10⁻⁶ M). TQEN binds Cd²⁺ with higher affinity than Zn²⁺.²³ The reduction in absolute Cd²⁺ affinity of 1b can be attributed to the formation of 6-membered chelate rings rather than the 5-membered chelate rings in the [Cd(TQEN)]²⁺ complex. The increase in chelate ring size decreases the binding affinity for large metal ions.³¹

Fig. 1 (a) UV-vis absorption and (b) fluorescence spectra of 34 μM 6-MeO-8-TQEN (1b) (λ_ex = 332 nm) in DMF–HEPES buffer (1 : 1) at 25 °C with Cd²⁺ between 0 and 1.7 mM. (c) Plot of integrated fluorescence intensity from 350 to 600 nm with increasing [Cd²⁺].

Fluorescence spectral change of 1b induced by other metal ions

Fig. 2 shows the fluorescence response of 1b to 20 equiv. of various metal ions. The fluorescence enhancement is Cd²⁺-specific although Zn²⁺ induces small fluorescence changes at 395 nm (I_Zn/I_Cd = 0.35). The lack of a fluorescence response to Pb²⁺ and Hg²⁺ enhances the value of 1b as a fluorescent probe for toxic metal ions. As shown in Fig. 2a, the addition of Zn²⁺ leads to an emission enhancement in the bands at 405 and 490 nm, the latter indicates the formation of an intramolecular excimer.²⁵ Due to this lower-energy fluorescence, the comparison based on the integrated emission is less specific to Cd²⁺ (F_Zn/F_Cd = 0.72). Since many Zn²⁺ probes respond to Cd²⁺, we are interested in developing a general probe design strategy to discriminate effectively between the two metal ions. In contrast to TQEN and 6-MeOTQEN, which exhibit a Zn²⁺-induced fluorescence enhancement of I_Zn/I₀ = 23 (I_Cd/I_Zn = 0.64) and I_Zn/I₀ = 11 (I_Cd/I₀ = 14, I_Cd/I_Zn = 1.29) respectively, the response of 1b is Cd²⁺-selective (I_Zn/I₀ = 9, I_Zn/I_Cd = 0.35). The fluorescence response suggests that six-membered chelate ring formation allows 1b to accommodate the larger Cd²⁺ ion. The alkoxy substitution further enhances the emission signal. The binding constant of Zn²⁺ with 1b was estimated to be K_d = (3.7 ± 0.3) × 10⁻⁴ M, which is significantly weaker binding than the Cd²⁺ binding. This difference is the plausible explanation for the Cd/Zn discrimination ability of 1b.

Fig. 2 (a) Fluorescence spectra and (b) the relative fluorescence intensity of 6-MeO-8-TQEN (1b) at 395 nm (I/I₀, filled bars) and integrated region from 350 to 600 nm (F/F₀, open bars) upon the addition of various metal ions. Emission was recorded with 20 equiv. of metal ion in DMF–HEPES buffer (1 : 1) at 25 °C. I₀ and F₀ are the emission intensity of free ligand.

UV-vis and fluorescence spectral changes of 2a–d induced by Cd²⁺

Since a larger binding pocket appears to enhance Cd²⁺-selectivity in 1b, we examined extending the alkyl linkers in 8-TQEN. In 8-TQOPEN (2b), the receptor forms four, six-membered chelate rings upon metal ion binding. The 8-quinolinol fluorophore was chosen to increase the intensity of fluorescence signal and because evidence suggests alkoxy linkers increase Cd²⁺-selectivity.¹⁵ Unfortunately, apo-2b in DMF–H₂O (1 : 1) emits intensely (Fig. 3, λ_ex = 321 nm; λ_em = 408 nm), due to inefficient PET quenching. The inefficient PET is probably related to the distance between the quinoline flurophores and the quenching ethylenediamine nitrogen lone pairs. Metal ions that typically quench fluorescent sensors such as Hg²⁺, Fe³⁺, Ag⁺ and Cu²⁺ also extinguish the emission of 2b in DMF–H₂O (1 : 1).

Fig. 3 (a) Fluorescence spectra of 34 μM 8-TQOPEN (2b) (λ_ex = 324 nm) in DMF–H₂O (1 : 1) at 25 °C in the presence of 2 equiv. of various metal ions. (b) The relative fluorescence intensity of 2b integrated from 350 to 600 nm upon the addition of 2 equiv. of various metal ions. I₀ is the emission intensity of free ligand.

To test the hypothesis that linker length leads to inefficient PET quenching in apo-8-TQOPEN (2b), 8-TQOEEN (2a) was prepared. The two probes are identical except 2b has a propylene linkage between the nitrogen atom and the quinoline, while 2a possesses an ethylene spacer. In contrast to the previous report,¹⁵ the emission of apo-2a is even greater than that of apo-2b (λ_ex = 321 nm; λ_em = 409 nm). Upon addition of 2 equiv. of metal ions, only an emission decrease analogous to the one observed with 2b was measured with quenching metal ions (Fig. 4). No significant response to Zn²⁺ or Cd²⁺ could be detected with this metal dose (2 equiv.).

Fig. 4 (a) Fluorescence spectra of 34 μM 8-TQOEEN (2a) (λ_ex = 324 nm) in DMF–H₂O (1 : 1) at 25 °C in the presence of 2 equiv. of various metal ions. (b) The relative fluorescence intensity of 2a integrated from 350 to 600 nm upon the addition of 2 equiv. of various metal ions. I₀ is the emission intensity of free ligand.

To further investigate the impact of six-membered chelate rings on sensor properties, we synthesized the propylenediamine derivatives 8-TQOEPN (2c) and 8-TQOPPN (2d). The propyl linker on the central diamine moiety introduces a six-membered chelate ring at the core of the complex. Although 2d, that possesses four propyleneoxy chains, exhibits considerable basal fluorescence intensity and responds to quenching metal ions like 2a and 2b (Fig. 5), the emission of 2c is reduced compared to the other structurally related derivatives (Fig. 6a). These optical properties suggest that the PET quenching is reasonably efficient. More importantly, 2c fluorescence increases in the presence of 2 equiv. of Cd²⁺ (Fig. 6b, gray bars) unlike related probes. Although a large excess of Cd²⁺ can induce a moderate fluorescence enhancement in 2a, the fluorescent intensity of [Cd(2c)]²⁺ is greater than that of apo-2a, 2b and 2d (Fig. 6b, filled bars). The composite results indicate that a combination of the PET and CHEF mechanisms is responsible for the signaling behavior. Furthermore, alkyl chain length impacts the PET efficiency of the apo-probe and the fluorescent response toward Cd²⁺.

Fig. 5 (a) Fluorescence spectra of 34 μM 8-TQOPPN (2d) (λ_ex = 324 nm) in DMF–H₂O (1 : 1) at 25 °C in the presence of 2 equiv. of various metal ions. (b) The relative fluorescence intensity of 2d integrated from 350 to 600 nm upon the addition of 2 equiv. of various metal ions. I₀ is the emission intensity of free ligand.

Fig. 6 Comparison of fluorescence properties of 2a–d. (a) Fluorescence spectra of 34 μM 2a–d in DMF–HEPES buffer (1 : 1) at 25 °C in the absence of metal ion. (b) Comparison of fluorescence intensity of 2a–d integrated from 350 to 600 nm in the absence (open bars), presence of 2 equiv. (gray bars), and presence of 20 equiv. of Cd²⁺ (filled bars). (c) Comparison of fluorescence enhancement factor (F/F₀) of 2a–d in the presence of 20 equiv. of Cd²⁺. F₀ is the integrated emission intensity of free ligand.

The Cd²⁺ binding of 2c was investigated further (Fig. 7). Upon excitation at 325 nm, emission at 423 nm increases gradually upon the addition of Cd²⁺. The fluorescence signal saturates at approximately 20 equiv. of metal ion with 19-fold emission enhancement (Fig. 7c). The fluorescence quantum yield of [Cd(2c)]²⁺ complex (ϕ = 0.31) is significantly higher than that of [Cd(1b)]²⁺. Assuming that 2c binds a single Cd²⁺ ion, the dissociation constant of 2c (K_d = 1.4 ± 0.2 × 10⁻⁴ M, Fig. S3†) indicates the lower affinity of 2c than 1b.

Fig. 7 (a) UV-vis absorption and (b) fluorescence spectra of 34 μM 8-TQOEPN (2c) (λ_ex = 325 nm) in DMF–HEPES buffer (1 : 1) at 25 °C with Cd²⁺ between 0 and 2.7 mM. (c) Plot of integrated fluorescence intensity from 350 to 600 nm with increasing [Cd²⁺].

Fluorescence spectral change of 2c induced by other metal ions

The fluorescence enhancement of 2c is Cd²⁺-specific. While Zn²⁺ induces a modest response (F_Zn/F₀ = 8.8 and F_Zn/F_Cd = 0.45), other metal ions do not modulate the emission (Fig. 8). No long wavelength emission upon Zn²⁺ binding is detected (Fig. 8a). The ability to discriminate Zn²⁺ and Cd²⁺ is comparable to probes with related structures.^14,15 The metal ion selectivity profiles of 2c and 1b (Fig. 2) are similar, despite the structural differences. From the Zn²⁺ titration curve, the binding constant of 2c with Zn²⁺ was estimated to be K_d = (4.0 ± 0.2) × 10⁻⁴ M (Fig. S4†). The binding affinity of 2c for Zn²⁺ and Cd²⁺ is nearly identical, although there is a slight preference for Cd²⁺. Although a slight difference in binding affinity could be partially responsible for the Cd/Zn selectivity of 2c, the fluorescence intensity of fully metalated 2c is metal-ion dependent. The difference in the emission intensity of the Zn²⁺ and Cd²⁺ complexes suggests a different coordination environment impacts sensor emission switching mechanism.

Fig. 8 (a) Fluorescence spectra and (b) the relative fluorescence intensity of 8-TQOEPN (2c) integrated from 350 to 600 nm upon the addition of various metal ions. Emission recorded with 20 equiv. of metal ion in DMF–HEPES buffer (1 : 1) at 25 °C. F₀ is the integrated emission intensity of free ligand.

pH effect on fluorescence intensity of 1b and 2c

The proton-induced emission changes of 1b and 2c were investigated in aqueous DMF solution (1 : 1) in the presence and absence of 5 equiv. of Cd²⁺ by adding either HCl or NaOH (Fig. 9). Due to the limited solubility of Cd(OH)₂, the complex was formed prior to the addition of NaOH solution by mixing 5 equiv. of CdCl₂ and 34 μM of the corresponding sensor. Both compounds show very narrow Cd²⁺ detection range (pH = 7–9 for 1b and pH = 7–8 for 2c) compared to related probes¹⁵ due to the weak metal binding affinity. Protonation of the ligand at the low pH, and the formation of Cd(OH)₂ at the high pH leads to demetalation of the sensor. No proton-induced fluorescence enhancement was observed for either 1b or 2c.

Fig. 9 Effect of pH on fluorescence intensity of 34 μM (a) 6-MeO-8-TQEN (1b) at 395 nm and (b) 8-TQOEPN (2c) at 423 nm in the absence (blue marks) and presence (red marks) of 5 equiv. of Cd²⁺ in DMF–H₂O (1 : 1) at 25 °C.

Conclusions

Two fluorescent sensors for Cd²⁺ inspired by TQEN, 6-MeO-8-TQEN (1b) and 8-TQOEPN (2c), have been developed. From the comparative analysis of 1b and the structurally related sensor 6-MeOTQEN, substituting an 8-quinolyl fluorophore for a 2-quinolyl one provides Cd²⁺ selectivity because of the receptor can accommodate larger metal ions more efficiently. Introducing 6-methoxy groups enhances the solubility of these compounds and red-shifts the excitation and emission wavelengths compared to TQEN. Of the four alkoxy-8-quinoline derivatives investigated, only 2c exhibited effective Cd²⁺-induced fluorescence enhancement (ϕ_Cd = 0.31). The other three compounds exhibited moderate fluorescence in their free forms and exhibited minimal responses to Cd²⁺; however, Hg²⁺, Ag⁺ and Cu²⁺ quench the emission. Since the weak metal binding affinity and narrow pH window for Cd²⁺ detection for 1b and 2c need to be improved, future work will focus on developing sensors with properties more suitable for cellular imaging using the general design strategies elucidated here.

Acknowledgements

This work was supported by the Research for Promoting Technological Seeds, JST, Adaptable and Seamless Technology Transfer Program through Target-driven R&D, JST, Grant-in Aid for Scientific Research from the MEXT, Japan and the Nara Women's University Intramural Grant for Project Research.

References

1. A. P. de Silva, H. Q. N. Gunaratne, T. Gunnlaugsson, A. J. M. Huxley, C. P. McCoy, J. T. Rademacher and T. E. Rice, Chem. Rev., 1997, 97, 1515–1566.
2. A. P. de Silva, D. B. Fox, A. J. M. Huxley, N. D. McClenaghan and J. Roiron, Coord. Chem. Rev., 1999, 185–186, 297–306.
3. E. L. Que, D. W. Domaille and C. J. Chang, Chem. Rev., 2008, 108, 1517–1549.
4. L. M. Hyman and K. J. Franz, Coord. Chem. Rev., 2012, 256, 2333–2356.
5. Z. Liu, W. He and Z. Guo, Chem. Soc. Rev., 2013, 42, 1568–1600.
6. R. D. Hancock, Chem. Soc. Rev., 2013, 42, 1500–1524.
7. B. Valeur and I. Leray, Coord. Chem. Rev., 2000, 205, 3–40.
8. C. J. Frederickson, E. J. Kasarskis, D. Ringo and R. E. Frederickson, J. Neurosci. Methods, 1987, 20, 91–103.
9. M. S. Nasir, C. J. Fahrni, D. A. Suhy, K. J. Kolodsick, C. P. Singer and T. V. O'Halloran, J. Bioinorg. Chem., 1999, 4, 775–783.
10. C. J. Fahrni and T. V. O'Halloran, J. Am. Chem. Soc., 1999, 121, 11448–11458.
11. E. Tomat and S. J. Lippard, Curr. Opin. Chem. Biol., 2010, 14, 225–230.
12. X. Peng, J. Du, J. Fan, J. Wang, Y. Wu, J. Zhao, S. Sun and T. Xu, J. Am. Chem. Soc., 2007, 129, 1500–1501.
13. T. Cheng, Y. Xu, S. Zhang, W. Zhu, X. Qian and L. Duan, J. Am. Chem. Soc., 2008, 130, 16160–16161.
14. X.-L. Tang, X.-H. Peng, W. Dou, J. Mao, J.-R. Zheng, W.-W. Qin, W.-S. Liu, J. Chang and X.-J. Yao, Org. Lett., 2008, 10, 3653–3656.
15. Z. Li, P. Xi, L. Huang, G. Xie, Y. Shi, H. Liu, M. Xu, F. Chen and Z. Zeng, Inorg. Chem. Commun., 2011, 14, 1241–1244.
16. L. Xue, G. Li, Q. Liu, H. Wang, X. Liu, C. Ding, S. He and H. Jiang, Inorg. Chem., 2011, 50, 3680–3690.
17. S. Goswami, K. Aich, S. Das, A. K. Das, A. Manna and S. Halder, Analyst, 2013, 138, 1903–1907.
18. A. M. Costero, S. Gil, J. Sanchis, S. Peransí, V. Sanz and J. A. G. Williams, Tetrahedron, 2004, 60, 6327–6334.
19. C. Lu, Z. Xu, J. Cui, R. Zhang and X. Qian, J. Org. Chem., 2007, 72, 3554–3557.
20. X. Zhou, P. Li, Z. Shi, X. Tang, C. Chen and W. Liu, Inorg. Chem., 2012, 51, 9226–9231.
21. Y. Mikata, Y. Sato, S. Takeuchi, Y. Kuroda, H. Konno and S. Iwatsuki, Dalton Trans., 2013, 42, 9688–9698.
22. Y. Mikata, Y. Nodomi, A. Kizu and H. Konno, Dalton Trans., 2014, 43, 1684–1690.
23. Y. Mikata, M. Wakamatsu and S. Yano, Dalton Trans., 2005, 545–550.
24. Y. Mikata, M. Wakamatsu, A. Kawamura, N. Yamanaka, S. Yano, A. Odani, K. Morihiro and S. Tamotsu, Inorg. Chem., 2006, 45, 9262–9268.
25. Y. Mikata, A. Yamanaka, A. Yamashita and S. Yano, Inorg. Chem., 2008, 47, 7295–7301.
26. Y. Mikata, K. Kawata, S. Iwatsuki and H. Konno, Inorg. Chem., 2012, 51, 1859–1865.
27. G. Weber, J. Riecke, W. Saenger and G. M. Sheldrick, Inorg. Chim. Acta, 1981, 54, L185–L186.
28. R. Bartzatt and L. Donigan, Lett. Drug Des. Discovery, 2004, 1, 78–83.
29. Y. Sakurai and M. Izumi, Pharm. Bull., 1953, 1, 297–301.
30. K. Schlögl and R. Schlögl, Monatsh. Chem., 1964, 95, 922–941.
31. W. Gan, S. B. Jones, J. H. Reibenspies and R. D. Hancock, Inorg. Chim. Acta, 2005, 358, 3958–3966.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details for preparation of new compounds, characterization data and Fig. S1–S26. See DOI: 10.1039/c4ra00010b

---